本文通过语音实验法和对比分析法，以四种高频二字组声调组合为例，对5名新西兰高中生汉语主题会话中产出的二字组声调进行偏误分析和动态发展研究。
实验研究的步骤为先将学生11-13年级主题会话的语料进行文字转录，对会话中出现的二字组按照声调组合模式进行标记和分类，统计出学生会话语流中出现频次最高的四种声调组合——阴平+去声、阳平+上声、上声+阴平和去声+去声，并使用praat语音软件对这四种组合模式中的常用语块进行切分和音高数据的提取，同时提取出母语者的相关数据和语图作为对照；然后通过T值公式对基频数据进行归一化处理，绘制出新西兰高中生和母语者的调值与调型分布图。然后根据分布图对学生的声调面貌进行细致描述，探讨新西兰学生的偏误表现和动态特征。
通过与母语者进行对比，得出新西兰学生二字组声调在语流中的偏误表现为：在阴平+去声组合主要是调域过窄和音高较低；阳平+上声组合中学生的偏误主要在于没有掌握阳平的上升调调型，而上声+阴平的偏误普遍在于后位阴平的调型和调值，同时由于学生没有掌握低调与高调之间的相对音高，因此学生这两类二字组声调的整体调域都较窄；在去声+去声组合中新西兰学生声调分布的空间较低，前位去声容易产生调类偏误变成其他调型，后位去声表现优于前位。总的来说学生在语流中的声调组合普遍表现为前位声调平调化，后位主要表现为缓降调，并且前后位声调之间的相对音高频差调整不准确。此外，通过对学生各自的声调格局进行描述，得出学生习得声调的过程不是线性上升的，而是具有循环往复的特征，在新西兰学生汉语学习过程中，声调格局的动态发展呈现出进步、磨蚀和石化等多种状态，并且每名学生的具体变化过程各不相同。
本文认为造成学生上述的偏误表现和动态特征的原因主要有三个，第一方面在于学生母语的语调和发音机制对汉语声调产出造成的负迁移；第二方面是由于语流中汉语声调本身相对音高的变化具有复杂性；第三方面来自于学生外部学习环境造成的不利影响，比如NCEA考试的导向与教师的教学策略等。文章最后对新西兰高中生的声调教学提出了相关建议，认为首先应当从理念和听觉感知上帮助学生建立起对声调的认知，其次要注重对学生进行调域拓展和声调定位的训练，同时应当抓住不同调类教学的重点，在语流中有的放矢进行声调教学，并且教师要有意识地延长学生语音学习的自觉期，在汉语教学的任何阶段都不能忽视声调训练。

In this paper, through the method of phonetic experiment and comparative analysis, taking four high-frequency two-syllable tones as examples, it makes research of the tone errors and dynamic development of five New Zealand senior high students in Chinese thematic conversation. 
The steps of experimental research are as follows. First, transcribe the corpus of students' thematic conversation from Year11 to Year13, mark and classify all the disyllabic words according to the tone combination mode, get the combination of four tones with the highest frequency in the conversation, which are T1+T4, T2+T3, T3+T1 and T4+T4. Praat is used to extract the pitch data of the most common two-syllable words in these four combination modes. Then, obtain the native speaker’s pronunciation data for comparison as well. In this paper, we use T-value formula to normalize the fundamental frequency, then draw the tone value and pattern distribution map of both the NZ students and the native speaker. According to the distribution map, the tone features of the students are described in detail to explore the errors and dynamic characteristics of the NZ senior high students.
By comparing with the native speakers’ tone map, we find that the two-syllable tone errors of NZ students in the natural speech flow are as follows. In the combination of the first tone and the fourth tone, both the tone pitch of first and rear is too low, so students’ tone range is much narrower than Chinese. In the combination of the second tone and the third tone, students’ main error is that they cannot pronounce the rising tone well, while the error in the combination of the third tone and the first tone is generally due to the tone value of the rear tone, which is flat and high. What’s more, because students have not mastered how to adjust relative pitch between low key and high key, so the overall tonal range of these two combinations is relatively narrow. The NZ students’ tonal distribution space is relatively low in the combination of the fourth tone and fourth tone, and the front tone is prone to have tonal errors then change to other tones, while the latter one is better than that of the former. 
In general, the tone combination of students in speech flow generally shows the flat tone as the anterior tone, and the slow descending tone as the posterior tone. Besides, the relative pitch difference between the front and rear tones is not adjusted accurately. In addition, through the description of the students’ individual tonal patterns, it is concluded that the process of students’ tonal acquisition is not a linear growth, but with a cyclical characteristic. In the process of students learning Chinese in senior high, the dynamic development of the tonal pattern shows progress, language attrition and fossilization, and the specific characteristics of each student’s development are different as well.
This paper holds that there are three main reasons for the above-mentioned errors and dynamic characteristics of students. The first is the negative transfer of the students’ mother tongue and pronunciation mechanism to the production of mandarin tones. The second is the complexity of the relative pitch of mandarin tone itself in speech flow; the third aspect comes from the adverse effects of the students' external learning environment, such as the orientation of the NCEA exam and the teaching strategies of the teacher. At the end of the paper, relevant suggestions are put forward for the tonal teaching of New Zealand high school students. It is believed that firstly, students should be helped to establish the cognition of tones from the perspective of concept and auditory perception. Secondly, the training of tonal expansion and tone positioning should be paid to the students. And teachers should also grasp the key point of different categories of tones’ teaching at the same time, target in mandarin tones teaching in speech flow as well. Moreover, teachers should prolong the conscious period of students' pronunciation learning consciously. All in all, pronunciation and other language elements are complementary and inseparable. Mandarin tones, as a part of Chinese, go through the whole stage of students' language learning process. When students master the correct pronunciation of tone, it can also promote the learning of other language elements.